Small‐Scale Seismic Monitoring of Varying Water Levels in Granular Media

Abstract

Core Ideas We tested the sensitivity of seismic measurements to water saturation variations. We used ultrasonic techniques to reproduce a small‐scale seismic acquisition setup. Measurements were performed on glass bead layers with varying water levels. Results of 3D numerical modelling were used to validate data at the dry state. Data difference trends between the dry and wet models match with the water level. Physical properties of soils in the vadose zone, and especially their water content, are characterized by strong spatial and temporal variations mostly driven by weather and anthropogenic activities. To understand this variability and help water resource management, seismic methods have been recently suggested as a complement to electrical and electromagnetic techniques. The simultaneous in situ estimation of pressure (P) and shear (S) wave velocities (VP and VS, respectively) and their ratio (VP/VS) offers novel perspectives for the monitoring of space and time variations of vadose zone mechanical properties. However, the seismic response in partially saturated and unconsolidated soils remains complex and deserves to be studied both theoretically and experimentally. In this study, we tested the sensitivity of seismic measurements (i.e., P‐wave travel times and surface‐wave phase velocities) to water saturation variations using controlled physical models at the laboratory scale. Ultrasonic techniques were used to reproduce a small‐scale seismic acquisition setup at the surface of glass bead layers with varying water levels. Travel times and phase velocity measurements obtained at the dry state were validated with both theoretical models and numerical simulations and serve as reference datasets. The increasing water level clearly affected the recorded wave field in both its phase and amplitude. In these cases, the collected data cannot yet be inverted in the absence of a comprehensive theoretical model for such partially saturated granular media. The differences in travel time and phase velocity observed between the dry and wet models show patterns that interestingly match the observed water level and depth of the capillary fringe, thus offering attractive perspectives for studying soil water content variations.

Estimating picking errors in near‐surface seismic data to enable their time‐lapse interpretation of hydrosystems

“…; Pasquet et al . ). Picked phase velocities are finally represented as a function of frequency and Xmid position of each extraction window (Fig.…”

“…2017), water level changes (Lu ; Pasquet et al . , ), permafrost thaw (Ajo‐Franklin et al . ) or deep fluid storage (Arts et al .…”

“…Pasquet et al . (2016a) use a similar approach to track water level changes at the laboratory scale in a well‐characterized unconsolidated granular material. Point‐by‐point data differences show inflection points at specific transition zones corresponding to the water and capillary fringe levels.…”

“…Time-lapse approaches for seismics have been recently used for a wide range of applications and at different scales to track changes of seismic signal and mechanical properties associated with climatic variabilities (Bergamo et al 2016a,b;Ikeda et al 2017), water level changes (Lu 2014;Pasquet et al 2015bPasquet et al , 2016a, permafrost thaw (Ajo-Franklin et al 2017) or deep fluid storage (Arts et al 2004;Chadwick et al 2010). In the meantime, passive seismics have also been developed and recently proved to be efficient in the continuous auscultation of subsurface layers at the 0.01% level (Lecocq et al 2017).…”

Parsimonious surface wave interferometry

“…As an example, seismic surfacewave techniques has been proposed for the detection of varying levels of the water table of shallow aquifers, heavy-oil in oil fields (Pasquet et al, 2015;Dubos-Sallee et al, 2015;Ikeda et al, 2016;Pasquet et al, 2016), or monitoring climate effects on earthworks (Bergamo et al, 2016a,b).…”